{"id":"ef02558a-85a5-41f0-9d98-aaee3e39dd1a","arxiv_id":"2505.13730","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"XRISM resolves the neutral iron K-alpha doublet in the low-luminosity AGN M81*, finding a narrow, unshifted line consistent with emission at r >= 2.7 x 10^4 GM/c^2, likely a remnant torus.","lead":"M81*, the closest low-luminosity black hole, was observed with the new X-ray calorimeter XRISM, which resolved the iron K-alpha line into two components. The line's narrow width places its origin far from the black hole, likely in a remnant torus, and gives a new upper limit on magnetic field strength there.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The r≥2.7×10^4 GM/c² radius and torus conclusion assume Keplerian broadening; no test separates Keplerian from turbulent/wind broadening.","rationale":"The Fe Kα measurement itself appears well supported: the doublet is seen, the centroid is consistent with systemic, and the width is resolved at roughly 9.9 eV FWHM, well above the Resolve line spread function. The problem is the physical conversion. Section 3.2 uses the Speith convolution, whose only broadening mechanism is Keplerian motion, so the fitted inner radius is determined by whatever velocity scale the line profile implies under that assumption. Nothing in the spectrum distinguishes a 460 km/s orbital shear from a 460 km/s turbulent velocity or a wind component, and the paper does not argue otherwise. In fact, the similarity to the Hα core FWHM (400 km/s) suggests a common, possibly non-Keplerian velocity field. Without an independent constraint on the velocity field, the lower limit is not a lower limit on radius but a lower limit conditional on the model. The title and abstract's 'torus' claim then rests on an interpretation, and the paper's own discussion remains appropriately hedged. Because the concern is substantial but not falsifying—the Keplerian assumption is plausible and could be checked—the appropriate handling is to keep the reader's CONDITIONAL verdict; my read does not move it. Agreement with the reader: the reader's weakest assumption is the same Keplerian-broadening issue.","tokens_in":17700,"tokens_out":6372,"duration_ms":65012,"concrete_test":"Re-fit the 1.7–10.7 keV Resolve spectrum of the neutral Fe Kα with a model that adds a Gaussian turbulent broadening kernel to the mytorus line while allowing the Speith inner radius to drop to small values (r~10 GM/c²). If the fit statistic changes by less than ΔC≈3 from the published Keplerian fit for σ_turb≈460 km/s, then the data cannot distinguish Keplerian from turbulent broadening, and the lower limit r≥2.7×10^4 GM/c² is model-dependent rather than data-required. To make the test decisive, also fit the reverse case: fix the radius at the published value and allow zero turbulence, and compare Akaike/Bayesian information; if both models have comparable evidence, the torus-radius claim should be downgraded to 'one possible geometry'.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The radius lower limit r≥2.7×10^4 GM/c², and with it the claim that Fe Kα traces a torus, rests on one untested assumption: that the measured broadening (FWHM=460+260−160 km/s) is Keplerian orbital motion around the black hole, as implemented in the Speith convolution (§3.2). The paper does not independently constrain the velocity field of the line-emitting gas. If the same line width is dominated by turbulence or a wind, the emitting region can lie at any radius, and the quoted lower limit no longer follows. This is not a contrived alternative: the paper itself notes that the Hα core has FWHM≈400 km/s and may trace the same geometry (§4.1), and it concedes in §4.5 that the data cannot distinguish emissivity profiles q=3 and q=2, so the geometry is not pinned down. The torus identification is therefore one of several allowed interpretations, and the ancillary support is incomplete: the MIR–NIR spectrum shows no clear torus excess (Mason et al. 2012), and the previously published Suzaku/NuSTAR reflection upper limit R≤0.1 (Young et al. 2018) is contradicted only by an unpublished, self-cited reanalysis (Miller et al. 2025, in prep.). The detection of the doublet and its centroid are credible; the load-bearing step is the width-to-radius mapping and the resulting geometric inference.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports XRISM/Resolve spectroscopy of M81*, a nearby low-luminosity AGN, over a 1.7–10.7 keV passband. The authors detect and resolve the neutral Fe K_alpha doublet into K_alpha,1 and K_alpha,2, measure a negligible velocity shift and a line width of FWHM = 460^{+260}_{-160} km/s, and use the Speith Keplerian convolution in SPEX to derive an inner emission radius of r >= 2.7 x 10^4 GM/c^2. On this basis they argue that the Fe K_alpha line likely traces a remnant torus rather than a disk or BLR. The paper also reports Fe XXV and Fe XXVI emission lines that can be fitted with either photoionization or collisional excitation, a marginal second ionized component redshifted by about 1600 km/s, and an upper limit on Zeeman splitting that translates into B <= 3.5 x 10^8 G in the line-emitting region. These results are discussed in the context of RIAF, MAD, and Fermi-bubble models.","tokens_in":17982,"tokens_out":6107,"duration_ms":61855,"significance":"If the central inference holds, this is the first resolved X-ray line measurement localizing cold gas at large radius in an LLAGN, and it would support the idea that Seyfert-like tori survive in a diminished form at L/L_Edd ~ 10^-5. The resolved Fe K_alpha doublet is a valuable new datum from a new instrument class, and the paper is unusually transparent about degeneracies: it presents both photoionization and collisional models, reports that q = 2 and q = 3 emissivity profiles cannot be distinguished, and notes where components are of marginal significance. Tables 1–3 give full parameter sets, and the background treatment for the solar-flare epoch is careful. The main limitation is that the headline radius and torus interpretation rest on an unvalidated assumption about the velocity field of the line-emitting gas, so the result is best regarded as a conditional but important measurement.","major_comments":[{"comment":"The lower limit r >= 2.7 x 10^4 GM/c^2 and the subsequent torus interpretation assume that the measured FWHM = 460^{+260}_{-160} km/s is dominated by Keplerian orbital motion, as implemented through the Speith convolution in SPEX. The data do not independently constrain the velocity field of the line-emitting gas; the same line width could be produced by turbulent broadening or an outflow at essentially any radius. The paper itself notes in Section 4.1 that the H_alpha core has FWHM ~ 400 km/s and may trace the same geometry, and admits in Section 4.5 that the data cannot distinguish between q = 2 and q = 3 emissivity profiles. I request an explicit test of the Keplerian assumption, for example fitting with an additive turbulent broadening term, comparing the full line-profile shape to the Speith prediction, or using an independent tracer to fix the radius. If no such test is possible, the radius and torus statements should be reframed as conditional on the Keplerian interpretation rather than presented as the primary result.","section":"Sections 3.2, 4.1, 4.5"},{"comment":"A key element of the torus interpretation is consistency with the published Suzaku/NuSTAR reflection upper limit R <= 0.1 (Young et al. 2018). The paper argues that this limit is overturned by a reanalysis of the same NuSTAR data, but the only reference is an unpublished work (Miller et al. 2025, in prep). This is not verifiable or reproducible from the present manuscript. Please include the NuSTAR reduction and fitting details, at least as an appendix, with spectra, model components, best-fit parameters, and the change in C-stat; alternatively, state clearly that the reflection constraint remains unresolved and adjust the strength of the torus claim accordingly.","section":"Section 4.1"},{"comment":"The quoted radius and its conservative lower limit depend on several fixed or weakly constrained parameters: N_H fixed to 1.6 x 10^24 cm^-2, emissivity index q fixed to 3, spin fixed to 0.7, and an inclination that is only weakly constrained. The paper reports that N_H cannot be constrained from the data and that q = 2 gives fully consistent radii, but it does not propagate the range of allowed N_H and q into the final r >= 2.7 x 10^4 GM/c^2 figure. Please provide a systematic error budget for the radius, or a table showing the dependence of the lower limit on the plausible range of these parameters, so that the abstract-level claim reflects this model dependence.","section":"Section 3.2, Tables 2 and 3"}],"minor_comments":[{"comment":"Please proofread the text for typographical errors, including \"Appenix A\" in Section 2, \"model-indepenent\" in Section 3.1, and the irregular spacing in the title \"Remnant T orus\".","section":"General"},{"comment":"The units of the Fe K_alpha normalization and of the pion/CIE emission-measure normalizations are not defined in the table notes; adding explicit units would make the tables self-contained.","section":"Tables 2 and 3"},{"comment":"The Zeeman formula taken from Inoue (2023) is written with a different normalization and parameter set than the earlier expression Delta E = 11.6 eV (B/10^9 G) in Section 3.1; please reconcile the two formulas or clarify the definition of each parameter.","section":"Section 4.4"},{"comment":"The second ionized component is repeatedly described as marginal, but the main text does not give a single quoted significance; please state the Delta C and number of additional free parameters for this component in both the photoionization and collisional models.","section":"Sections 3.3 and 3.4"}],"recommendation":"major_revision","confidential_remarks":"This is a valuable observation and the resolved Fe K_alpha doublet is a clear step forward for LLAGN studies. The main concern is not the quality of the data but the strength of the geometric inference: the radius and torus conclusion rest on an untested Keplerian assumption and on an unpublished reanalysis. If the authors can add a test of the velocity-field assumption and include the NuSTAR reanalysis details, I would be happy to support publication; as it stands, the central claim needs either additional support or a more conditional framing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The important thing here is the measurement, not the title. XRISM/Resolve gives the first calorimeter spectrum of an LLAGN and clearly resolves the neutral Fe K-alpha doublet in M81*. The narrow width (FWHM = 460 km/s) and the resulting radius lower limit of r >= 2.7e4 GM/c^2 are a genuine advance over the old Chandra grating limit of r >= 1e3 GM/c^2. The Zeeman-based upper limit on the magnetic field is a nice extra, though it is too weak to meaningfully constrain MAD models, as the authors themselves admit. The spectral fits are careful: explicit models, 1-sigma errors, and an honest treatment of the photoionization-versus-collisional degeneracy in the ionized lines. This is solid work with real archival value.\n\nThe soft spot is exactly what the stress-test note flags. The radius lower limit is only as good as the Keplerian broadening assumption built into the Speith convolution. The paper does not independently constrain the velocity field of the line-emitting gas, and a turbulent or wind-like broadening of ~460 km/s could place the emitting gas at almost any radius. The paper actually concedes much of this in Section 4.5, where it admits that an emissivity index of q=2 gives fully consistent radius constraints, and in Section 4.1 where it notes that the H-alpha core has a similar width and may trace the same geometry. The torus interpretation is therefore one of several allowed options, not the unique conclusion the title implies. The supporting reflection-fraction argument also leans on an unpublished NuSTAR reanalysis (Miller et al. 2025, in prep.) that reverses the earlier Young et al. upper limit; that is a real citation-pattern concern, though it does not affect the line measurement itself.\n\nI do not think the Keplerian assumption is fatal here. It is the standard way to interpret resolved line widths in AGN, and the alternative explanations are mentioned rather than ignored. But the paper overreaches at the level of the abstract and title, and a referee should push back on that.\n\nWho is this for? X-ray astronomers working on LLAGN, AGN structure, and Fe K spectroscopy. It deserves a serious referee and will probably be cited as the reference measurement for the Fe K line in M81*, regardless of whether the torus interpretation survives. The honest conclusion is: accept the measurement, treat the torus claim with caution, and require the authors to either tone down the interpretation or provide stronger evidence that the broadening is genuinely Keplerian.","headline":"A real and useful XRISM measurement of the Fe K-alpha line in M81*, but the torus claim is an interpretation built on the untested Keplerian-broadening assumption, not a direct detection.","tokens_in":18779,"tokens_out":1526,"would_cite":true,"duration_ms":16658,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The resolved iron K-alpha line in M81* is narrow and nearly at rest, placing the emitting gas at least 27,000 gravitational radii from the black hole and pointing to a remnant torus.","keywords":["low-luminosity AGN","M81*","iron K-alpha line","XRISM/Resolve","torus","accretion disk truncation","radiatively inefficient accretion flow","X-ray spectroscopy"],"falsifier":"Fit the resolved Fe K$\\alpha$ line with a model that allows a free non-Keplerian velocity component, such as turbulence plus outflow, and compare the inferred radius: if a turbulent width of order 460 km/s gives an equally good fit with an emission radius below about $10^3\\,GM/c^2$, the quoted lower limit is not unique. Alternatively, a deep observation that detects Fe K$\\alpha$ flux variability on a timescale shorter than the light-crossing time across $2.7\\times 10^4\\,GM/c^2$ would place the gas much closer to the black hole and overturn the torus interpretation.","tokens_in":17492,"feed_emoji":"🛰️","tokens_out":10323,"duration_ms":93488,"temperature":0.7,"pith_summary":"Using the XRISM/Resolve microcalorimeter, this paper resolves the neutral iron K-$\\alpha$ line of M81*, the nearest and brightest low-luminosity active galactic nucleus, into its two fine-structure components. The line is narrow (FWHM about 460 km/s) and shows no significant Doppler shift, so if its broadening is Keplerian the gas must sit at least $2.7\\times 10^4\\,GM/c^2$ from the black hole. The authors interpret this as the signature of a remnant molecular torus, a leftover of a more active phase, and support that reading with a small but nonzero reflection fraction. The same spectrum shows He-like Fe XXV and H-like Fe XXVI lines, consistent with either photoionized or collisionally excited plasma, with a marginally significant redshifted component near 1600 km/s. The result matters because up to 40% of local galaxies host low-luminosity active nuclei, and it has been unclear whether the standard Seyfert disk-torus geometry survives at an Eddington ratio near $10^{-5}$.","feed_headline":"A narrow iron line reveals a remnant torus around M81*","feed_subtitle":"The line width puts the gas at 27,000 gravitational radii, restoring the torus to low-luminosity AGN.","key_machinery":"The load-bearing object is the neutral Fe K$\\alpha$ doublet, cleanly separated by the Resolve calorimeter. The radius claim is produced by the Speith convolution in SPEX, which takes a reflection line function, here the mytorus model, and reprocesses it under the assumption that the emitting gas moves on Keplerian orbits around a spinning black hole; the inner and outer radii, inclination, and emissivity index are fit parameters. With emissivity index $q=3$ and spin $a=0.7$ fixed, the fits give inner radii around $4\\text{--}6\\times 10^4\\,GM/c^2$, and the paper quotes a conservative lower limit of $2.7\\times 10^4\\,GM/c^2$. The same line doublet supplies the magnetic-field bound through the Zeeman splitting relation $\\Delta E = 11.6\\,\\mathrm{eV}(B/10^9\\,\\mathrm{G})$. The pion and CIE models in SPEX are used to attribute the Fe XXV and Fe XXVI emission to photoionized or collisionally excited gas, respectively.","core_discovery":"The paper's central claim is that the neutral Fe K$\\alpha$ line in the XRISM/Resolve spectrum of M81* is resolved into K$\\alpha,1$ and K$\\alpha,2$ components with a negligible velocity shift and modest broadening, FWHM $= 460^{+260}_{-160}$ km/s, and that a Keplerian model of that broadening places the line-forming region at $r \\geq 2.7\\times 10^4\\,GM/c^2$ for the inclinations allowed by ultraviolet modeling. At that radius the gas is far outside the inner accretion flow and the broad-line region, and the paper concludes that the line most plausibly traces the inner wall of a remnant torus with a low covering factor. This would be the first direct X-ray indication that a Seyfert-like torus survives in a low-luminosity active nucleus accreting at about $10^{-5}$ Eddington. The paper also reports Fe XXV and Fe XXVI emission consistent with a wind or possibly with the early stage of a galactic bubble, and an upper limit of $B \\leq 3.5\\times 10^8$ G on the magnetic field in the line region from the absence of extra Zeeman splitting.","pith_inferences":["My inference, not the paper's: the torus identification hinges on the Keplerian assumption, and a turbulence- or wind-dominated velocity field of a few hundred km/s would collapse the radius lower limit; a dedicated line-profile study with higher signal-to-noise could distinguish these cases.","A testable extension would be to measure infrared dust reverberation lags in M81* at radii near $10^4\\text{--}10^5\\,GM/c^2$; agreement with the Fe K$\\alpha$ radius would independently confirm the remnant-torus picture.","The marginal redshifted component near 1600 km/s is close to the free-fall velocity at roughly $8\\times 10^4\\,GM/c^2$ quoted in the paper; stacking multiple XRISM visits could turn this marginal feature into a kinematic constraint on the accretion flow.","The paper's Zeeman argument implies that for brighter low-luminosity AGN, Zeeman splitting will usually be hidden inside broader line profiles; continuum polarization or jet-power scaling may be more informative for MAD magnetic fields than line-splitting searches."],"forward_implications":["If the torus reading is right, the canonical AGN components - a hot inner disk, a broad-line region, and a dusty torus - do not switch off at $L/L_{\\rm Edd}\\approx 10^{-5}$; they persist in reduced, partially depleted form.","The radius lower limit of $2.7\\times 10^4\\,GM/c^2$ implies that the inner disk, if present, is far from the innermost stable circular orbit, matching the truncated-disk expectation of radiatively inefficient accretion flow and magnetically arrested disk models.","The narrow Fe K$\\alpha$ line is kinematically distinct from the broad ionized Fe lines, so future deep XRISM exposures should be able to separate torus, wind, and broad-line-region components by velocity alone.","The Zeeman upper limit of $3.5\\times 10^8$ G at $2.7\\times 10^4\\,GM/c^2$ is too weak to constrain magnetically arrested disks; line-width splitting in low-luminosity AGN is therefore unlikely to be an incisive probe of MAD magnetic fields.","If the 1600 km/s redshifted Fe XXVI component is real, its velocity and the lack of a blue-shifted counterpart favor a far-side wind or a natal outflow bubble, both testable with line variability."],"supporting_citations":[{"why":"supplies the exact fitting template, mytorus plus Speith, used for the narrow Fe K alpha line and the NGC 4151 torus-radius comparison.","marker":"XRISM Collaboration et al. 2024"},{"why":"previous joint Suzaku and NuSTAR analysis that placed the Fe K alpha region at $r\\geq 10^3\\,GM/c^2$ and reported a reflection upper limit that this paper re-examines.","marker":"Young et al. 2018"},{"why":"the radiatively inefficient accretion flow model for M81* whose truncated-disk prediction is the main theoretical frame, including its explicit caveat about the black hole mass.","marker":"Quataert & Narayan 1999"},{"why":"provides the mytorus line function that models the neutral Fe K complex and its Compton shoulder.","marker":"Murphy & Yaqoob 2009"},{"why":"provides the Keplerian convolution used to convert line broadening into an emission-radius constraint.","marker":"Speith et al. 1995"},{"why":"supplies the black hole mass and the 14-degree inclination used to restrict the radius fits.","marker":"Devereux et al. 2003"},{"why":"earlier Chandra grating detection of Fe XXV and Fe XXVI lines that defines the baseline wind and redshift interpretation.","marker":"Young et al. 2007"},{"why":"the MHD wind model whose predicted blue-shifted emission is compared against the XRISM limits.","marker":"Shi et al. 2021"}],"fun_headline_variants":["XRISM reveals remnant torus in LLAGN M81*","Iron line pins torus at 27,000 GM/c2","First torus seen in LLAGN M81*","XRISM catches M81* with a Seyfert-like torus"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conversion from the measured 460 km/s line width to the lower limit of $2.7\\times 10^4\\,GM/c^2$ assumes that the broadening comes from Keplerian orbital motion around the black hole; if turbulence or a wind dominates the velocity field, the radius and torus interpretation lose their support.","fun_headline_variants_meta":{"raw":{"variants":["XRISM reveals remnant torus in LLAGN M81*","Iron line pins torus at 27,000 GM/c2","First torus seen in LLAGN M81*","XRISM catches M81* with a Seyfert-like torus"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001375,"raw_usage":{"total_tokens":5694,"prompt_tokens":1191,"completion_tokens":4503,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":807,"completion_tokens_details":{"reasoning_tokens":4430}},"tokens_in":807,"tokens_out":4503,"duration_ms":30810,"temperature":1.0,"reasoning_tokens":4430,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:10:32.834942+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fit the resolved Fe K$\\alpha$ line with a model that allows a free non-Keplerian velocity component, such as turbulence plus outflow, and compare the inferred radius: if a turbulent width of order 460 km/s gives an equally good fit with an emission radius below about $10^3\\,GM/c^2$, the quoted lower limit is not unique. Alternatively, a deep observation that detects Fe K$\\alpha$ flux variability on a timescale shorter than the light-crossing time across $2.7\\times 10^4\\,GM/c^2$ would place the gas much closer to the black hole and overturn the torus interpretation.","supporting_citations":[],"review_version":1}